In May of last month, OpenAI unveiled the GPT-4o model and demonstrated an impressive Voice Mode that enables users to engage in human-like conversations with ChatGPT. Initially, OpenAI aimed to begin rolling out the new Voice Mode to a limited group of ChatGPT Plus users by the end of June. However, OpenAI has now announced that it requires “an additional month to meet our standards for launch.”
The AI startup explains that it is enhancing the GPT-4o model for voice interactions and improving its “capability to identify and reject specific content.” OpenAI is also working on optimizing its infrastructure to support real-time responses and reduce latency, in preparation for scaling the Voice Mode feature to millions of users.
OpenAI plans to initiate a phased deployment starting with an alpha preview for a small user group one month later. By the end of this year, all ChatGPT Plus users are expected to gain access to the new Voice Mode. OpenAI has refrained from specifying an exact timeline, emphasizing that the rollout will be influenced by user feedback.
Additionally, OpenAI is preparing to introduce features like video and screen sharing, showcased during the event. Notably, Scarlett Johansson disclosed that OpenAI CEO Sam Altman has been in contact with her regarding potentially lending her voice for the new Voice Mode on ChatGPT.
Following Johansson’s decline, OpenAI developed a ‘Sky’ voice that closely resembled her character from the movie ‘Her’. After the actress raised concerns, OpenAI promptly removed the ‘Sky’ voice from ChatGPT. Moving forward, users can anticipate discovering the range of new voices available once the enhanced Voice Mode is officially launched.
0 Comments